Not signed in (Sign In)

# Start a new discussion

## Not signed in

Want to take part in these discussions? Sign in if you have an account, or apply for one below

• Sign in using OpenID

## Site Tag Cloud

Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.

• CommentRowNumber1.
• CommentAuthorUrs
• CommentTimeApr 29th 2012
• CommentRowNumber2.
• CommentAuthorTobyBartels
• CommentTimeMay 3rd 2012

I expanded on this in a couple of ways.

• CommentRowNumber3.
• CommentAuthorUrs
• CommentTimeMay 3rd 2012

Thanks, nice!

I have created a stub for pointless topology just so as to keep the link from being grey.

• CommentRowNumber4.
• CommentAuthorzskoda
• CommentTimeMay 8th 2012

Let $S$ be a topological subspace $S\subset \mathbf{R}^n$ of a Cartesian space.

I find “of a Cartesian space” both superfluous and confusing as it is already written that it is a subset of $\mathbf{R}^n$ and the topology there is standard for most of the Earth’s mathematicians. On the other hand, I have never seen in the literature outside of ncat community expression “Cartesian space” for $\mathbf{R}^n$ and when I read this then I start clicking to find out what it is, as it looks to me that maybe a greater generality is hidden somewhere in the statement. So I was confused when reading the statement.

• CommentRowNumber5.
• CommentAuthorTobyBartels
• CommentTimeMay 8th 2012

You’re not missing anything. I think that the main for saying the words “Cartesian space” is to have something to put the link on. (I forget if Urs or I put those words in.)

The only really standard term for a Cartesian space is the symbol “$\mathbb{R}^n$”, but it’s nice to have a word too, and the concept is closely associated with Descartes. Some people say “Euclidean space”, but that has another meaning which is more appropriate for Euclid (who used no coordinates).

• CommentRowNumber6.
• CommentAuthorUrs
• CommentTimeMay 8th 2012
• (edited May 8th 2012)

Yes, I agree with Toby. On the $n$Lab I happily link even to terms that I find dead-standard. Because somebody may come who does not know the term.

Maybe better would be in this case here to have the symbol “$\mathbb{R}^N$” be hyperlinked. But that’s not really possible, is it. And not really nicer, either, I find.

You know, it even gives me a little pain each time that an entry says “$let n \in \mathbb{Z}$” without linking to integer. Because, optimally, every single technical point would be linked to its explanation. Who knows, next time the 10-year old genius-to-be comes reading our $n$Lab pages and wants to know what $\mathbb{R}^N$ denotes, or what $\mathbb{Z}$ denotes. In any case, links are what a wiki is for. Books without hyperlinks we had enough. And it slowed mankind down for long enough.

• CommentRowNumber7.
• CommentAuthorzskoda
• CommentTimeMay 9th 2012
• (edited May 9th 2012)

Look, it scares me when I see a text with lots of unfamiliar words like Cartesian space. What about people who are much less new-definition and theory friendly, as most mathematicians are ? If a word is not necessary it can be linked in brackets, then one knows it is not important to know it. It looked to me that one is pointing to a more general context of some beasts called Cartesian spaces, to which the classical Heine-Borel is generalized. If Cartesian were in brackets rather than a part of the linear statement, I would neglect it and maybe even learn without clicking. Clicking may be easy for some, but to me, who does more than half of the work from bad connections, sometimes a new loading of a page may mean a couple of minutes freeze of a browser. And yes, I think we should be able to link eventually somehow from symbols (Urs, you were always more of a technology optimist than I, don’t give up this time), even from formulas.

• CommentRowNumber8.
• CommentAuthorMike Shulman
• CommentTimeMay 9th 2012

What about “Let S be a topological subspace of a Cartesian space, i.e. $S\subseteq \mathbb{R}^n$ for some $n\in\mathbb{N}$”? The “i.e.” clues the reader that nothing more is being said by the first phrase than the second, hence “Cartesian space” must refer to $\mathbb{R}^n$.

• CommentRowNumber9.
• CommentAuthorzskoda
• CommentTimeMay 9th 2012
• (edited May 9th 2012)

That is much better. Even more familiar textbook phrase is also “subspace $S$ of real $n$-dimensional space $\mathbf{R}^n$.

• CommentRowNumber10.
• CommentAuthorMirco Richter
• CommentTimeMay 9th 2012
• (edited May 9th 2012)

Guess the most popular name is just “coordinate space” for the $\mathbb{R}^n$’s, because that’s the term you lerned in ’undergraduate’ . Obviously the reason are the base isomorphisms from any vector space into an apropriate $\mathbb{R}^n$.

• CommentRowNumber11.
• CommentAuthorUrs
• CommentTimeMay 9th 2012

a text with lots of unfamiliar words like Cartesian space

I was not aware that you count “Cartesian space” as an “unfamilar word”!

• CommentRowNumber12.
• CommentAuthorzskoda
• CommentTimeMay 9th 2012
• (edited May 9th 2012)

Hmm, I am unfamiliar with “coordinate space” terminology either. Are you then unfamiliar in a similar vain with “real $n$-dimensional space” ?

Urs: I know that it was appearing in a context of smooth spaces, but after not seeing it for a while, I am never sure which one is which (what is the restriction on morphisms, do we count smooth structure or not, and is it for presheaves or for original test spaces). So at first I thought that it is a synonym for a smooth space, thus that the intended theorem is a generalization.

• CommentRowNumber13.
• CommentAuthorMirco Richter
• CommentTimeMay 9th 2012

“real n-dimensional space” is not unique to $\mathbb{R}^n$ but points to any vector space isomorphic to $\mathbb{R}^n$.

• CommentRowNumber14.
• CommentAuthorzskoda
• CommentTimeMay 9th 2012
• (edited May 9th 2012)

Then I would say “a real $n$-dimensional vector space”. Maybe one should emphasize “the” ?

• CommentRowNumber15.
• CommentAuthorUrs
• CommentTimeMay 9th 2012
• (edited May 9th 2012)

But wait, “Cartesian space” is a dead standard term no? Here this is taught in high school. Analytic geometry.

Whether you take the maps betwen them to be continuous, smooth or linear depends, no matter how you call it.

• CommentRowNumber16.
• CommentAuthorMike Shulman
• CommentTimeMay 9th 2012

I didn’t learn “coordinate space” or “Cartesian space” in high school or undergraduate. We called it “$\mathbb{R}^n$”.

• CommentRowNumber17.
• CommentAuthorUrs
• CommentTimeMay 9th 2012

But you say “cartesian coordinates” in schood, don’t you?

• CommentRowNumber18.
• CommentAuthorTobyBartels
• CommentTimeMay 9th 2012

Before reading Mike #8, I reacted to Zoran #7 by editing things in a similar line.

@Urs: Although I very much like the term “Cartesian space”, I learnt it only from you. Maybe it’s a German thing?

• CommentRowNumber19.
• CommentAuthorMike Shulman
• CommentTimeMay 9th 2012

Yes, “cartesian coordinates” sometimes, but just as often “rectangular coordinates” or just “coordinates”. And not “cartesian space”.

• CommentRowNumber20.
• CommentAuthorUrs
• CommentTimeMay 9th 2012

Okay, I see. I wasn’t aware of that. I thought I was using a common term.

• CommentRowNumber21.
• CommentAuthorUrs
• CommentTimeApr 19th 2017
• (edited Apr 19th 2017)

I have spelled out a classical proof of the Heine-Borel theorem, here

Add your comments
• Please log in or leave your comment as a "guest post". If commenting as a "guest", please include your name in the message as a courtesy. Note: only certain categories allow guest posts.
• To produce a hyperlink to an nLab entry, simply put double square brackets around its name, e.g. [[category]]. To use (La)TeX mathematics in your post, make sure Markdown+Itex is selected below and put your mathematics between dollar signs as usual. Only a subset of the usual TeX math commands are accepted: see here for a list.

• (Help)